Fabricated building construction quality monitoring system based on Internet of Things

Abstract

The invention provides a fabricated building construction quality monitoring system based on the Internet of Things, and relates to the technical field of building construction quality monitoring, and the system comprises a sensor module which is used for obtaining various quality related data in a construction process in real time, and the quality related data comprises a structural member installation position, connection strength, concrete pouring quality, and environment temperature and humidity; the data transmission module is used for transmitting various quality related data of the sensor module to the data processing module; and the data processing module is used for receiving and processing the data of the data transmission module, performing data analysis, anomaly detection and quality evaluation, and generating a construction quality evaluation result. Through real-time data acquisition, intelligent analysis processing and remote monitoring management, the construction quality monitoring efficiency and accuracy are effectively improved, intelligent control over construction equipment is achieved, the safety of the construction process is ensured, meanwhile, a user can conveniently obtain construction information and alarm prompts in time, and the construction quality is improved. And the convenience and response speed of construction management are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of building construction quality monitoring, and in particular to an Internet of Things-based prefabricated building construction quality monitoring system. Background Art

[0002] Traditional manual inspection methods mainly rely on the inspectors' vision and experience to judge construction quality. However, due to limited human energy and vision, as well as the large construction sites and large number of components of large-scale prefabricated building projects, it is difficult for inspectors to conduct comprehensive and detailed inspections of all construction areas.

[0003] For example, in a large-scale prefabricated building project, due to the vast construction area, inspectors may only be able to inspect some key areas, while ignoring construction quality issues in some hidden or marginal areas, resulting in the construction quality of these areas not being effectively guaranteed.

[0004] Although the sampling detection method can reflect the construction quality to a certain extent, it is difficult to comprehensively and accurately evaluate the quality of the entire construction process due to the limitation of the number of samples. At the same time, the manual recording of data may also have the possibility of errors and omissions. For example, when testing the installation strength of components, the traditional method may only randomly select some components for testing, and it is impossible to monitor the installation strength of all components in real time and accurately.

[0005] In addition, when inspectors record data manually, negligence or fatigue may cause inaccurate data recording, affecting the evaluation results of construction quality. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide an Internet of Things-based prefabricated building construction quality monitoring system, which realizes real-time monitoring and intelligent evaluation of construction quality, and improves construction efficiency and building quality.

[0007] In order to solve the above technical problems, the technical solution of the present invention is as follows: The prefabricated building construction quality monitoring system based on the Internet of Things includes: The sensor module is used to obtain various quality-related data during the construction process in real time, including the installation position of structural parts, connection strength, concrete pouring quality, and ambient temperature and humidity; A data transmission module, used to transmit various quality-related data of the sensor module to the data processing module; A data processing module is used to receive and process data from the data transmission module, perform data analysis, anomaly detection and quality assessment, and generate construction quality assessment results; Intelligent equipment control module, used to automatically adjust the working parameters of construction equipment or stop the operation of faulty equipment according to the construction quality assessment results; An alarm module is used to detect whether the construction quality assessment result meets the preset standard, and send out an alarm signal when the construction quality assessment result does not meet the preset standard; The user terminal module is used to display construction quality assessment results and alarm information, and supports remote control.

[0008] Furthermore, the data from the data transmission module is received and processed, data analysis, anomaly detection and quality assessment are performed, and construction quality assessment results are generated, including: Extract features related to construction quality assessment from the data in the data transmission module, including the installation position deviation of structural parts, the value of connection strength, the uniformity index of concrete pouring, and the range of changes in ambient temperature and humidity; According to the preset anomaly detection threshold, the extracted features are detected, the existing anomalies or abnormal patterns are identified and marked, and anomaly detection results are formed; Based on historical data, a construction quality assessment model is constructed, and the extracted features and feature sets of anomaly detection results are input into the quality assessment model for calculation and reasoning to obtain the construction quality assessment results.

[0009] Furthermore, according to a preset anomaly detection threshold, the extracted features are detected, and the existing anomalies or abnormal patterns are identified and marked to form an anomaly detection result, including: The preset anomaly detection model is trained and learned through the isolation forest algorithm to obtain a trained anomaly detection model; The extracted features, including the installation position deviation of the structural parts, the numerical value of the connection strength, the uniformity index of the concrete pouring, and the ambient temperature and humidity, are passed as input data to the trained anomaly detection model; After training, the anomaly detection model analyzes the input features and calculates an anomaly score for each data point; According to the anomaly score, a threshold is set, and the anomaly score of each data point is compared with the threshold. When the anomaly score of a data point exceeds the threshold, the data point is marked as an anomaly point or an abnormal pattern; The data points marked as abnormal are aggregated to form the anomaly detection results, including the feature values, occurrence time and abnormal type information of the abnormal points.

[0010] Furthermore, the calculation formula for the anomaly score of a data point is: ; in, Represents data points Anomaly score of Indicates the size of the data set; , represents the scaling factor; represents the number of trees in the isolation forest; Represents the index of the tree; Represents data points In the The path length in the tree; Indicates The weight of each tree; Indicates Tree-specific adjustment factor; represents a positive number; Indicates adjustment The coefficient of influence; Indicates the offset; Represents parameters related to tree depth.

[0011] Furthermore, a construction quality assessment model is constructed based on historical data, and the extracted features and feature sets of anomaly detection results are input into the quality assessment model for calculation and reasoning to obtain construction quality assessment results, including: Extract construction quality-related features from historical construction data, including installation position deviation of structural components, connection strength, concrete pouring uniformity, and ambient temperature and humidity; The random forest model is used as the construction quality assessment model, and the construction quality assessment model is trained using the training set in the historical data. During the training process, the parameters and hyperparameters of the construction quality assessment model are adjusted to obtain a trained quality assessment model. The extracted features and feature sets of anomaly detection results are input into the trained construction quality assessment model for analysis, and the construction quality assessment results are obtained through calculation, including the construction quality score, qualified or unqualified judgment, and identification of potential problems.

[0012] Furthermore, the calculation formula for the construction quality score is: ; in, A score indicating the quality of construction; represents a constant; Indicates the total number of features; , The index representing the feature; Indicates The value of a feature; Indicates The type coefficient of the feature; Indicates The standardized coefficients of the features; Indicates The standardized coefficients of the features; represents the adjustment parameter; represents the base of natural logarithms; Indicates The minimum value of the eigenvalue among all samples; Indicates The maximum value of the eigenvalue among all samples; Indicates and The interaction coefficient between the features; Indicates The value of a feature; Indicates time; Indicates time The weight of time; Indicates time The characteristic value when ; Represents the time decay rate.

[0013] Furthermore, according to the construction quality assessment results, the working parameters of the construction equipment are automatically adjusted or the operation of the faulty equipment is stopped, including: Presetting a threshold for construction quality assessment, and comparing the actual construction quality assessment result with the threshold to obtain a comparison result between the assessment result and the threshold; According to the comparison between the evaluation result and the threshold value, it is determined whether to adjust the equipment working parameters or stop the faulty equipment; For equipment that needs to adjust working parameters, the adjustment plan is determined based on the comparison between the evaluation results and the threshold, including the adjustment of speed, force and temperature parameters; for faulty equipment, the equipment stops working immediately; Generate corresponding control instructions based on the decision to adjust the plan or stop work, and send the control instructions to the construction equipment; The construction equipment automatically adjusts its own working parameters to the specified values ​​or stops working immediately according to the control instructions, and starts the fault handling process, including alarm and self-diagnosis.

[0014] Furthermore, according to the decision to adjust the plan or stop working, a corresponding control instruction is generated and sent to the construction equipment, including: receiving a decision signal to adjust the plan or stop working, and determining an initial radius of the trust region; Define an objective function to evaluate the quality of control instructions; Searching for new control instructions in the trust region and evaluating the impact of the new control instructions on the objective function to obtain an evaluation result; Compare the new control instruction with the current control instruction to obtain a comparison result, and determine whether to update the control instruction according to the comparison result; Continuously search for new control instructions in the trust domain, evaluate each new instruction, and determine whether to update the current control instruction based on the evaluation results until the maximum number of iterations is reached, generate the final control instruction, and send the final control instruction to the construction equipment.

[0015] Furthermore, the calculation formula of the objective function is: ; in, represents the objective function; , represents the weight coefficient; Indicates the actual construction location; Indicates the expected construction location; Indicates the base tolerance; Indicates the tolerance adjustment factor; Indicates basic working power; Indicates device parameters Linear influence coefficient on power; Indicates equipment parameters that affect energy consumption; Indicates device parameters The secondary influence coefficient on power; Indicates base time; Indicates the amplitude of periodic changes in energy consumption; represents the angular frequency; Indicates time.

[0016] Further, whether the construction quality assessment result meets the preset standard is detected, and when the construction quality assessment result does not meet the preset standard, an alarm signal is issued, including: Compare the construction quality assessment results with the preset standards to obtain comparative results; Analyze the comparison results to determine whether the construction quality assessment results meet the preset standards. If the assessment results meet the standards, continue monitoring; if not, enter the alarm signal processing process; In the alarm signal processing process, an alarm signal is generated, including log record, email notification, SMS reminder, sound alarm or light indication.

[0017] The above solution of the present invention includes at least the following beneficial effects: The sensor module can obtain various quality-related data in real time during the construction process, including the installation position of structural parts, connection strength, concrete pouring quality, ambient temperature and humidity, etc., to ensure the comprehensiveness and accuracy of the data. These data provide a solid foundation for subsequent data analysis, anomaly detection and quality assessment, and improve the accuracy of construction quality monitoring.

[0018] The data transmission module transmits various quality-related data from the sensor module to the data processing module, realizing automatic data transmission. The data processing module receives and processes the data, performs data analysis, anomaly detection and quality assessment, and generates construction quality assessment results. This automated process improves the efficiency of data processing, reduces manual intervention, and shortens the monitoring cycle.

[0019] The intelligent equipment control module automatically adjusts the working parameters of the construction equipment or stops the operation of the faulty equipment according to the construction quality assessment results. This intelligent control method reduces the error rate of human operation, ensures the normal operation of the construction equipment, and further guarantees the construction quality.

[0020] The alarm module detects whether the construction quality assessment results meet the preset standards. When the construction quality assessment results do not meet the preset standards, an alarm signal is immediately issued. This enables managers or operators to respond quickly to construction quality issues and take timely measures to correct them, thus avoiding the expansion and deterioration of the problem.

[0021] The user terminal module displays construction quality assessment results and alarm information, and supports remote control. This allows managers to understand the construction quality of the construction site anytime and anywhere, and make decisions and adjustments in a timely manner. At the same time, the remote control function also improves the flexibility and convenience of management. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of an Internet of Things-based prefabricated building construction quality monitoring system provided by an embodiment of the present invention.

[0023] Figure 2 The present invention provides an embodiment of the Internet of Things-based prefabricated building construction quality monitoring system, which detects extracted features according to a preset anomaly detection threshold, identifies and marks existing anomalies or anomaly patterns, and forms a flow chart of anomaly detection results. DETAILED DESCRIPTION

[0024] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0025] like Figure 1 As shown, an embodiment of the present invention proposes an assembled building construction quality monitoring system based on the Internet of Things, including: The sensor module 11 is used to obtain various quality-related data in the construction process in real time, including the installation position of structural parts, connection strength, concrete pouring quality, and ambient temperature and humidity; A data transmission module 12, used to transmit various quality-related data of the sensor module to the data processing module; The data processing module 13 is used to receive and process the data from the data transmission module, perform data analysis, anomaly detection and quality assessment, and generate construction quality assessment results; The intelligent equipment control module 14 is used to automatically adjust the working parameters of the construction equipment or stop the operation of the faulty equipment according to the construction quality evaluation results; An alarm module 15 is used to detect whether the construction quality assessment result meets the preset standard, and send out an alarm signal when the construction quality assessment result does not meet the preset standard; The user terminal module 16 is used to display the construction quality assessment results and alarm information, and supports remote control.

[0026] In an embodiment of the present invention, various quality-related data in the construction process, such as the installation position of structural parts, connection strength, concrete pouring quality, ambient temperature and humidity, etc., are obtained in real time and accurately. Provide a comprehensive and objective data basis. Improve the accuracy and reliability of construction quality monitoring and ensure the controllability and stability of the construction process. Efficiently and stably transmit various quality-related data obtained by the sensor module 11 to the data processing module 13. Realize real-time update and synchronization of data to ensure that the data processing module 13 can obtain the latest construction quality data in a timely manner. Reduce delays and losses in the data transmission process and improve the accuracy and reliability of data transmission.

[0027] Receive and process the data transmitted by the data transmission module 12, perform data analysis, anomaly detection and quality assessment. Perform in-depth mining and analysis of construction quality data through algorithms and models to generate objective construction quality assessment results. Improve data processing efficiency and accuracy, and provide timely and accurate decision-making basis for the intelligent device control module 14 and the alarm module 15. According to the construction quality assessment results generated by the data processing module 13, automatically adjust the working parameters of the construction equipment or stop the operation of the faulty equipment. Realize intelligent control of construction equipment and reduce human intervention and errors. Improve the operating efficiency and safety of construction equipment and ensure the smooth progress of the construction process.

[0028] Check whether the construction quality assessment results meet the preset standards. When the construction quality assessment results do not meet the preset standards, an alarm signal will be issued immediately. Timely remind managers or operators to pay attention to construction quality problems and prompt them to take prompt measures to correct them. Avoid the expansion and deterioration of construction quality problems and ensure the stability and reliability of construction quality.

[0029] Display construction quality assessment results and alarm information, so that managers can understand the construction quality of the construction site at any time. Support remote control function, so that managers can monitor and manage the construction process anytime and anywhere. Improve management flexibility and convenience, reduce management costs and improve management efficiency.

[0030] In a preferred embodiment of the present invention, various quality-related data in the construction process are obtained in real time, including the installation position of the structural parts, the connection strength, the quality of the concrete pouring, and the ambient temperature and humidity; and various quality-related data of the sensor module are transmitted to the data processing module, which may include: At the construction site of prefabricated buildings, various sensors are deployed reasonably according to construction quality and monitoring needs.

[0031] Structural component installation position sensor: used to monitor the installation position of structural components in real time to ensure that the structural components are accurately installed according to design requirements.

[0032] Connection strength sensor: installed at the connection of structural parts, used to monitor the strength of the connection in real time to ensure that the connection is firm and reliable.

[0033] Concrete pouring quality sensor: embedded in the concrete pouring area to monitor the concrete pouring quality, such as density, uniformity, etc.

[0034] Environmental temperature and humidity sensors: deployed in key areas of the construction site to monitor the ambient temperature and humidity of the construction site in real time and provide environmental parameters for construction quality control.

[0035] Each sensor collects various quality-related data during the construction process in real time according to the preset sampling frequency.

[0036] The collected data includes digital signals, analog signals, etc., which are processed and converted according to the type and working principle of the sensor. The collected raw data is preprocessed, including denoising, filtering, calibration, etc., to improve the accuracy and reliability of the data. The data is timestamped to record the collection time of the data.

[0037] According to the actual situation of the construction site, a stable and reliable data transmission network is built. The appropriate data transmission protocol is selected to ensure accurate transmission and effective reception of data. The sensor module sends the pre-processed data to the data processing module through the data transmission network according to the selected data transmission protocol. The data processing module receives the data from the sensor module and performs data verification and error correction to ensure the integrity and accuracy of the data. The data processing module stores the received data in the database. Establish a data management system to classify, index and back up the data to improve the accessibility and security of the data.

[0038] In a preferred embodiment of the present invention, receiving and processing data from the data transmission module, performing data analysis, anomaly detection and quality assessment, and generating a construction quality assessment result may include: Extract features related to construction quality assessment from the data in the data transmission module, including the installation position deviation of structural parts, the value of connection strength, the uniformity index of concrete pouring, and the range of changes in ambient temperature and humidity; According to the preset anomaly detection threshold, the extracted features are detected, the existing anomalies or abnormal patterns are identified and marked, and anomaly detection results are formed; Based on historical data, a construction quality assessment model is constructed, and the extracted features and feature sets of anomaly detection results are input into the quality assessment model for calculation and reasoning to obtain the construction quality assessment results.

[0039] In the embodiment of the present invention, the data stream sent by the data transmission module is continuously monitored through the data interface.

[0040] When new data arrives, it is immediately received and stored in a temporary buffer. The received raw data is parsed to identify key information such as the installation position of structural components, connection strength, concrete pouring quality, and ambient temperature and humidity. For the installation position of structural components, the deviation between the actual installation position and the designed position is calculated. For the connection strength, the value provided by the sensor is directly read. For the quality of concrete pouring, the uniformity index of concrete is calculated by analyzing the sensor data of the pouring area. For the ambient temperature and humidity, the range of temperature and humidity changes is recorded.

[0041] According to construction specifications and historical experience, a series of anomaly detection thresholds are preset, such as the maximum allowable value of installation position deviation, the minimum requirement for connection strength, the standard range of concrete pouring uniformity, and the suitable range of ambient temperature and humidity. The extracted features are compared with the preset thresholds. If a feature value exceeds the threshold range, it is marked as an abnormal point or abnormal pattern. Anomaly detection results are generated, and the specific information and occurrence time of the abnormal point are recorded.

[0042] Collect and organize historical construction data, including data from normal construction and abnormalities. Use machine learning algorithms to train historical data and build a construction quality assessment model. The model can learn the difference between normal construction and abnormal construction and generate assessment rules. The extracted features and the feature set of the anomaly detection results are passed as input to the construction quality assessment model. Based on the input feature set, calculations and reasoning are performed according to the learned assessment rules. The construction quality assessment results are output to indicate whether the current construction status is normal or what problems exist.

[0043] Assume that you are monitoring a prefabricated building construction site: The data received from the sensor includes the installation position of the structural member D. , connection strength is 100N / mm², uniformity index of concrete pouring area is 85%, ambient temperature is 25℃, and humidity is 60%. The installation position deviation of structural member D is calculated to be 2mm (compared with the design position), and the connection strength, uniformity index, ambient temperature and humidity are extracted as features. After checking the extracted features, it is found that the connection strength of 100N / mm² is within the normal range (threshold is 90-120N / mm²), but the uniformity index of 85% is slightly lower than the standard range (threshold is 88-92%), so it is marked as an abnormal point. Generate anomaly detection results and record the specific information and occurrence time of the uniformity index anomaly. Input the extracted features and anomaly detection results into the construction quality assessment model. The model performs calculations and inferences based on the input feature set and the evaluation rules learned from historical data. The model outputs the construction quality assessment results, indicating that there are minor problems with the quality of concrete pouring, and recommends further inspection and processing.

[0044] By extracting and analyzing key features of the construction process in real time, it is possible to accurately identify abnormal points or abnormal patterns, providing strong support for construction quality control. It is possible to continuously monitor various indicators of the construction site, promptly discover and report abnormal situations, and enable managers to quickly take corrective measures. By building a construction quality assessment model, it is possible to automatically evaluate the construction process, reduce manual intervention and subjectivity of judgment, and improve construction efficiency and quality. The generated construction quality assessment results provide managers with objective and accurate data support, helping them make more informed decisions and optimize the construction process and resource allocation.

[0045] In another preferred embodiment of the present invention, the extracted features are detected according to a preset anomaly detection threshold, and the existing abnormal points or abnormal patterns are identified and marked to form an abnormality detection result, which may include: The preset anomaly detection model is trained and learned through the isolation forest algorithm to obtain a trained anomaly detection model; The extracted features, including the installation position deviation of the structural parts, the numerical value of the connection strength, the uniformity index of the concrete pouring, and the ambient temperature and humidity, are passed as input data to the trained anomaly detection model; After training, the anomaly detection model analyzes the input features and calculates an anomaly score for each data point. The calculation formula for the anomaly score of a data point is: ; in, Represents data points Anomaly score of Indicates the size of the data set; , represents the scaling factor; represents the number of trees in the isolation forest; Represents the index of the tree; Represents data points In the The path length in the tree; Indicates The weight of each tree; Indicates Tree-specific adjustment factor; represents a positive number; Indicates adjustment The coefficient of influence; Indicates the offset; Represents parameters related to tree depth.

[0046] According to the anomaly score, a threshold is set, and the anomaly score of each data point is compared with the threshold. When the anomaly score of a data point exceeds the threshold, the data point is marked as an anomaly point or an abnormal pattern; The data points marked as abnormal are aggregated to form the anomaly detection results, including the feature values, occurrence time and abnormal type information of the abnormal points.

[0047] In an embodiment of the present invention, historical construction data is collected, including structural member installation position deviations under normal and abnormal conditions, connection strength values, concrete pouring uniformity indicators, and environmental temperature and humidity characteristics.

[0048] Preprocess the data, such as cleaning, denoising, and normalization, to ensure the quality and consistency of the data.

[0049] The Isolation Forest algorithm is selected as the basis of the anomaly detection model. The preprocessed historical data is used as a training set and input into the Isolation Forest algorithm. The algorithm builds multiple isolation trees (a variant of the random forest), each of which attempts to isolate a data point, thereby learning the normal distribution of the data. After multiple iterations of training, the trained anomaly detection model is obtained, which can identify data points that are significantly different from the normal distribution.

[0050] Extract features such as the installation position deviation of structural parts, the value of connection strength, the uniformity index of concrete pouring, and the ambient temperature and humidity from the real-time data stream. The extracted features are passed as input data to the trained anomaly detection model. The model receives the input data and prepares to calculate the anomaly score for each data point. Using the trained anomaly detection model, the anomaly score is calculated for each data point based on the principle of the isolation forest algorithm. Among them, Represents data points The higher the score, the more abnormal the data point is. Indicates the size of the dataset used to tune the baseline for anomaly scores. and is a scaling factor used to adjust the range of anomaly scores. Represents the number of trees in the isolation forest. The more trees there are, the more fully the model learns the data. is the index of the tree, indicating the tree currently being calculated. Represents data points In the The length of the path in the tree. The longer the path, the more likely the data point is to be isolated, that is, the more abnormal it is. It is The weight of each tree is used to adjust the impact of different trees on the anomaly score. It is A tree-specific tuning factor used to further fine-tune each tree's contribution to the anomaly score. is a positive number used to prevent the denominator from being zero or too small. Yes Adjustment The coefficient of influence. is the offset used to adjust the overall level of anomaly score. It is a parameter related to the depth of the tree. The deeper the tree, the longer the path length and the greater the contribution to the anomaly score.

[0051] According to historical data and actual needs, set a threshold for anomaly scores. Compare the anomaly score of each data point with the threshold. When the anomaly score of a data point exceeds the threshold, mark it as an anomaly point or an abnormal pattern. Summarize all data points marked as abnormal. Record the characteristic value, occurrence time, and anomaly type information of each anomaly point. Organize the summarized anomaly point information into an anomaly detection result report.

[0052] The report includes detailed information about the anomaly, the frequency of the anomaly, possible causes, and recommended treatment measures.

[0053] Assume that you are monitoring a prefabricated building construction site and have trained an anomaly detection model using the Isolation Forest algorithm. A set of features is extracted from the real-time data stream: the installation position deviation of structural member H is 3mm, the connection strength is 95N / mm², the concrete pouring uniformity index is 80%, the ambient temperature is 28℃, and the humidity is 55%. This set of features is used as input data and passed to the trained anomaly detection model. The model calculates the anomaly score of the data point based on the input features. Assume that the calculated anomaly score is 0.85 (this is an assumed value). The set anomaly score threshold is 0.8. Since 0.85 > 0.8, the data point is marked as an anomaly. The data points marked as anomalies are summarized to generate an anomaly detection result report. The report points out that there are anomalies in the installation position deviation of structural member H and the concrete pouring uniformity index, and further inspection and processing are recommended.

[0054] By training the anomaly detection model with the Isolation Forest algorithm, it is possible to more accurately identify abnormal points or abnormal patterns in the construction process. The calculation formula of the anomaly score takes into account multiple factors, such as the size of the data set, the number of trees, the path length, etc., making anomaly detection more comprehensive and detailed. Real-time monitoring of various indicators during the construction process, timely detection and reporting of abnormal situations, so that managers can quickly take corrective measures.

[0055] The anomaly detection result report provides detailed anomaly information and recommended treatment measures, which helps managers quickly locate and solve problems. Through anomaly detection, potential problems in the construction process can be discovered in a timely manner, such as excessive deviation in the installation position of structural parts and insufficient connection strength, thereby avoiding possible quality accidents and safety hazards. The training and learning process of the anomaly detection model can be continuously optimized and improved, so that construction quality and safety can be continuously improved.

[0056] In another preferred embodiment of the present invention, a construction quality assessment model is constructed based on historical data, and the extracted features and feature sets of anomaly detection results are input into the quality assessment model for calculation and reasoning to obtain a construction quality assessment result, which may include: Extract construction quality-related features from historical construction data, including installation position deviation of structural components, connection strength, concrete pouring uniformity, and ambient temperature and humidity; The random forest model is used as the construction quality assessment model, and the construction quality assessment model is trained using the training set in the historical data. During the training process, the parameters and hyperparameters of the construction quality assessment model are adjusted to obtain a trained quality assessment model. The extracted features and feature sets of anomaly detection results are input into the trained construction quality assessment model for analysis, and the construction quality assessment results are obtained through calculation, including the construction quality score, qualified or unqualified judgment, and identification of potential problems.

[0057] In an embodiment of the present invention, a database or file system storing historical construction data is accessed to obtain detailed records of construction processes including structural member installation, concrete pouring, etc. The data is cleaned to remove duplicate, missing or obviously erroneous data points to ensure data quality. Historical data is analyzed to identify features closely related to construction quality, such as structural member installation position deviation, connection strength, concrete pouring uniformity, and ambient temperature and humidity during construction. Data processing techniques (such as statistical analysis) are used to extract specific feature values ​​from these records.

[0058] The random forest model is selected as the basic model for construction quality assessment because it has good classification and regression performance and can handle high-dimensional data and nonlinear relationships. A part of the historical data is divided as a training set to ensure that the training set contains enough samples to cover various possible construction situations and quality levels. The training set is input into the random forest model to start the training process. During the training process, the parameters of the model (such as the number of trees, maximum depth, minimum number of sample splits, etc.) and hyperparameters (such as the selection of random seeds, feature selection methods, etc.) are continuously adjusted to optimize the performance of the model. The training effect of the model is evaluated by cross-validation and other techniques to ensure that the model is neither overfitting nor underfitting. After multiple iterations of training and optimization, the trained construction quality assessment model is obtained, which can accurately assess the construction quality based on the input feature values. The real-time extracted features (such as the installation position deviation of the structural parts in the current construction process, the connection strength, etc.) and the anomaly detection results (such as the feature values ​​of the anomaly points, the anomaly type, etc.) are combined into a feature set. The model receives the input data and starts to analyze and calculate. The input features are classified and regressed through multiple decision trees in the random forest, and the results of each tree are combined to obtain the final construction quality assessment result. The construction quality assessment results are obtained from the model, including the construction quality score (such as a percentage system or a graded system), the qualified or unqualified judgment, and the identification of potential problems (such as which characteristic values ​​are abnormal or may cause quality problems).

[0059] Suppose you are conducting a construction quality assessment on a bridge project that is under construction.

[0060] A set of features is extracted from the real-time data stream: the installation position deviation of the pier structure is 2mm, the connection strength is 100N / mm², the concrete pouring uniformity index is 90%, the ambient temperature is 25℃, and the humidity is 60%. At the same time, the anomaly detection results are also obtained, indicating that there is an anomaly in the uniformity of concrete pouring in a certain part of the pier. The extracted features and anomaly detection results are combined into a feature set and input into the trained construction quality assessment model. The input feature set is analyzed and calculated, and the influence of each feature value and the anomaly detection result is comprehensively considered. The construction quality assessment results are obtained from the model: the construction quality score is 85 points (out of 100), which is judged to be qualified, but it is pointed out that the uniformity anomaly of the pier concrete pouring may affect the long-term durability of the bridge, and further inspection and processing are recommended.

[0061] By evaluating the construction quality through the random forest model, the impact of multiple feature values ​​and abnormal detection results can be comprehensively considered to improve the accuracy of the evaluation. The training and optimization process of the model can continuously learn and adapt to new construction conditions and quality levels to ensure the timeliness of the evaluation results. Real-time feature extraction and construction quality evaluation can timely discover potential problems in the construction process and provide decision-making support for managers. The qualified or unqualified judgment and identification of potential problems in the evaluation results help managers quickly locate problems and take measures to correct them. Through construction quality evaluation, it can ensure that the construction process meets the design requirements and specifications, and improve construction quality and safety. The recommended treatment measures in the evaluation results can provide specific guidance for construction personnel, helping them to improve construction methods and techniques, and improve construction quality and efficiency.

[0062] In another preferred embodiment of the present invention, the calculation formula for the construction quality score is: ; in, A score indicating the quality of construction; represents a constant; Indicates the total number of features; , The index representing the feature; Indicates The value of a feature; Indicates The type coefficient of the feature; Indicates The standardized coefficients of the features; Indicates The standardized coefficients of the features; represents the adjustment parameter; represents the base of natural logarithms; Indicates The minimum value of the eigenvalue among all samples; Indicates The maximum value of the eigenvalue among all samples; Indicates and The interaction coefficient between the features; Indicates The value of a feature; Indicates time; Indicates time The weight of time; Indicates time The characteristic value when ; Represents the time decay rate.

[0063] In the embodiment of the present invention, all parameters and variables required in the initialization formula, including constants , total number of features , Feature Index , , eigenvalue , type coefficient , standardized coefficient and , Adjustment parameters ,time , time weight , time decay rate Etc. Make sure all parameters and variables are set correctly and the required feature values ​​can be obtained from the data source.

[0064] Calculate the basic score part, that is The partial score takes into account the average of all features and a constant , as the basis for scoring.

[0065] For each feature , calculate its type influence part, that is, .

[0066] This part of the score takes into account the type coefficient of the feature , standardized coefficient And the eigenvalue The normalized position relative to its minimum and maximum values, smoothed by a logistic function.

[0067] Calculate the interaction part between features, that is, This part of the score takes into account the interaction coefficients between features And the eigenvalue and The product of .

[0068] For each time point , the calculation time influence part, that is, This part of the score takes into account the time weight , Time point The eigenvalue of and the time decay rate , and time decay is processed by exponential function. The basic score part, feature type influence part, feature interaction part and time influence part are added together to get the final construction quality score. .

[0069] By calculating the construction quality score, multiple factors can be fully considered, including the average value of the characteristics, type influence, interaction, and time influence. This makes the scoring result more accurate and comprehensive, and can more truly reflect the actual situation of construction quality. Parameters and weights (such as , , , , The construction quality score can be flexibly adjusted according to the actual situation. This enables the scoring model to adapt to different construction projects and scenarios, improving the versatility and applicability of the model. By calculating the construction quality score, a clear basis can be provided for the quantitative evaluation of construction quality. This helps managers and construction personnel to understand the pros and cons of construction quality more intuitively, so as to take corresponding measures to improve and enhance it. The calculation process of the construction quality score can serve as a feedback mechanism to help construction personnel promptly identify problems and deficiencies in the construction process. By continuously adjusting and optimizing construction methods and techniques, the construction quality score can be gradually improved, thereby promoting the continuous improvement and enhancement of construction quality.

[0070] In a preferred embodiment of the present invention, automatically adjusting the working parameters of the construction equipment or stopping the operation of the faulty equipment according to the construction quality evaluation result may include: Presetting a threshold for construction quality assessment, and comparing the actual construction quality assessment result with the threshold to obtain a comparison result between the assessment result and the threshold; According to the comparison between the evaluation result and the threshold value, it is determined whether to adjust the equipment working parameters or stop the faulty equipment; For equipment that needs to adjust working parameters, the adjustment plan is determined based on the comparison between the evaluation results and the threshold, including the adjustment of speed, force and temperature parameters; for faulty equipment, the equipment stops working immediately; Generate corresponding control instructions based on the decision to adjust the plan or stop work, and send the control instructions to the construction equipment; The construction equipment automatically adjusts its own working parameters to the specified values ​​or stops working immediately according to the control instructions, and starts the fault handling process, including alarm and self-diagnosis.

[0071] In an embodiment of the present invention, a series of construction quality assessment thresholds are preset in the system, and these thresholds are determined according to historical data, industry standards or project requirements, and are used to measure the qualified standards of construction quality. The actual construction quality assessment results are obtained from the construction quality assessment model, which include scoring, qualified or unqualified judgments, etc. The actual construction quality assessment results are compared with the preset thresholds to determine whether the assessment results meet or exceed the threshold standards. According to the comparison process, the comparison results between the assessment results and the thresholds are obtained, such as "qualified", "unqualified" or a specific degree of deviation. The comparison results between the assessment results and the thresholds are analyzed to determine whether the construction quality meets the requirements. If the assessment results show that the construction quality is unqualified or there is a deviation, it is determined whether it is necessary to adjust the working parameters of the equipment or stop the faulty equipment according to the preset rules. For adjustable equipment, consider adjusting the working parameters to improve the construction quality; for faulty equipment, decide to stop its work immediately.

[0072] Identify the equipment that needs to adjust the working parameters. According to the degree of deviation between the evaluation results and the threshold, formulate a specific adjustment plan, including adjusting the speed, force and temperature of the equipment. The adjustment plan aims to optimize the working state of the equipment to improve the construction quality. For faulty equipment, stop working immediately. Identify the faulty equipment through the evaluation results. Immediately issue a stop working instruction to the faulty equipment to ensure that it does not continue to operate. Generate specific control instructions based on the adjustment plan or the decision to stop working. The control instruction includes the specific value of adjusting the working parameters of the equipment or the command to stop working. The control instruction is sent to the corresponding construction equipment through the communication system. The construction equipment receives the control instruction. According to the content of the control instruction, the construction equipment automatically adjusts its own working parameters to the specified value or stops working immediately. For the faulty equipment that stops working, the construction equipment starts the fault handling process, including issuing an alarm signal and performing self-diagnosis to identify and repair the fault.

[0073] Suppose you are monitoring the performance of a concrete mixer.

[0074] The threshold of concrete mixing quality is preset, such as the uniformity index must reach more than 90%. The mixing quality assessment result of the current batch of concrete is obtained from the construction quality assessment model, and the uniformity index is 85%.

[0075] The actual evaluation results were compared with the preset threshold value, and it was found that the uniformity index was lower than the threshold value. It was decided to adjust the working parameters of the mixer to improve the mixing quality. According to the degree of deviation between the evaluation results and the threshold value, an adjustment plan was formulated, including increasing the mixing speed, adjusting the mixing intensity and maintaining a suitable mixing temperature. Specific control instructions were generated, including adjusting the parameter values ​​of the mixing speed, intensity and temperature, and the instructions were sent to the mixer. After receiving the control instructions, the mixer automatically adjusted its own working parameters to the specified values ​​and began to mix with the new parameters. If the mixer fails to work properly due to a fault, the evaluation results will identify the fault of the mixer. Immediately send a stop command to the mixer. After receiving the stop command, the mixer stops working immediately and starts the fault handling process, including issuing an alarm signal and performing self-diagnosis.

[0076] By automatically adjusting the working parameters of construction equipment, the working status of the equipment can be optimized in real time to ensure that the construction quality is always within the qualified range, thereby improving the overall construction quality. It can promptly detect and stop the operation of faulty equipment to avoid more serious damage or safety accidents caused by continued operation of the faulty equipment. At the same time, the fault handling process can be started to reduce the downtime of faults. Automatically adjusting the working parameters of equipment and promptly stopping faulty equipment can reduce manual intervention and waiting time, improve construction efficiency, and ensure that the construction progress is carried out as planned. By real-time monitoring and adjusting the working status of equipment, potential safety hazards can be discovered and handled in a timely manner, enhancing the safety of the construction process and protecting the lives and property of construction personnel.

[0077] In another preferred embodiment of the present invention, according to the adjustment plan or the decision to stop working, generating corresponding control instructions and sending the control instructions to the construction equipment may include: receiving a decision signal to adjust the plan or stop working, and determining an initial radius of the trust region; Define an objective function to evaluate the quality of control instructions; Searching for new control instructions in the trust region and evaluating the impact of the new control instructions on the objective function to obtain an evaluation result; Compare the new control instruction with the current control instruction to obtain a comparison result, and determine whether to update the control instruction according to the comparison result; Continuously search for new control instructions in the trust domain, evaluate each new instruction, and determine whether to update the current control instruction based on the evaluation results until the maximum number of iterations is reached, generate the final control instruction, and send the final control instruction to the construction equipment.

[0078] In an embodiment of the present invention, an adjustment plan or a decision signal to stop working is received from an upper-level decision system through internal communication or an external input interface. According to preset rules or historical experience, an initial radius of a trust region is determined. This radius determines the range of parameter space that can be explored when searching for new control instructions. According to the control target (such as construction quality, equipment efficiency, energy consumption, etc.), a suitable objective function type is selected, such as minimizing error, maximizing efficiency, etc. In the trust region, new control instructions are searched according to a certain search strategy (such as random search). The new control instruction is substituted into the objective function, and the evaluation result of the objective function is calculated, that is, the potential impact of the new control instruction on the control target. The new control instruction is compared with the control instruction currently in use to evaluate the performance difference between the two on the objective function. If the new control instruction performs better than the current instruction on the objective function and meets other constraints (such as safety, feasibility, etc.), it is decided to update the control instruction.

[0079] Repeat the above process of searching, evaluating, comparing and updating to continuously search for better control instructions within the trust domain. When the preset maximum number of iterations is reached, the search stops and the final control instruction is generated. The final control instruction is sent to the construction equipment through the communication interface to make a decision to adjust the plan or stop working.

[0080] Suppose that the excavation speed of an excavator is being controlled to optimize construction efficiency. A signal is received from the upper decision-making system, requiring the excavator's excavation speed to be adjusted to improve construction efficiency. Based on historical experience and the performance parameters of the excavator, an initial radius of a trust region is determined, such as ±10% of the current speed. An objective function is defined to maximize construction efficiency. The objective function considers multiple factors such as excavation speed, excavation volume, and energy consumption. Randomly search for new excavation speed instructions within the trust region, such as increasing the speed by 5%. Substitute the new speed instruction into the objective function to evaluate its impact on construction efficiency. Compare the new speed instruction with the current speed instruction and find that the new instruction performs better on the objective function. Continue to iterate the search, constantly try new speed instructions, and evaluate their impact on the objective function. After multiple iterations, a final speed instruction is found that maximizes construction efficiency while also meeting the requirements of safety and feasibility. The final speed instruction is sent to the excavator, and the excavator adjusts the excavation speed according to the instruction.

[0081] Through trust region search and objective function evaluation, the final control instructions can be found more accurately, improving the control accuracy and efficiency of construction equipment. The trust region method limits the search range and avoids system instability or failure caused by excessive parameter adjustment. At the same time, through iterative search and evaluation, the final solution can be gradually approached to enhance the robustness of the system. Automatic search and evaluation of control instructions reduces manual intervention and trial and error costs and improves the automation level of the construction process. By maximizing the objective function of construction efficiency, the utilization of construction resources can be optimized, such as reducing energy consumption, increasing excavation volume, etc., thereby reducing construction costs.

[0082] In another preferred embodiment of the present invention, the calculation formula of the objective function is: ; in, represents the objective function; , represents the weight coefficient; Indicates the actual construction location; Indicates the expected construction location; Indicates the base tolerance; Indicates the tolerance adjustment factor; Indicates basic working power; Indicates device parameters Linear influence coefficient on power; Indicates equipment parameters that affect energy consumption; Indicates device parameters The secondary influence coefficient on power; Indicates base time; Indicates the amplitude of periodic changes in energy consumption; represents the angular frequency; Indicates time.

[0083] In the embodiment of the present invention, the weight coefficient is confirmed , , which determine the relative importance of the two sub-goals (position deviation and energy consumption) in the overall goal. and expected construction location , and how the deviation between them is calculated. Understanding Basic Tolerances , Tolerance adjustment factor , basic working power , Linear influence coefficient of equipment parameters on power , secondary influence coefficient , basic time , the amplitude of periodic changes in energy consumption , angular frequency and time The specific meaning and value of parameters such as

[0084] Calculate the deviation between the actual construction position and the expected construction position, that is, For each deviation term, calculate . Sum the squares of all deviation items and take the square root to get the value of the position deviation sub-target. , calculate the linear part of the operating power and the secondary part .

[0085] The basic working power Add the linear and quadratic parts to get the total operating power. Multiply by the base time , get the basic energy consumption. Considering the periodic changes in energy consumption, calculate , and multiply it by the basic energy consumption to get the value of the energy consumption sub-goal. Multiply the position deviation sub-goal and the energy consumption sub-goal by their weight coefficients respectively. and . Add the two weighted sub-objectives to get the value of the total objective function.

[0086] By comprehensively considering the two sub-goals of position deviation and energy consumption, the construction process can be optimized more accurately to ensure that the construction position is accurate and the energy consumption is inefficient. By searching and evaluating different control instructions, the final combination of construction parameters can be found, thereby improving construction efficiency and reducing construction time. The periodic changes in energy consumption are taken into account in the optimization process, and equipment parameters can be adjusted according to actual conditions to reduce energy consumption costs and improve energy utilization efficiency. Control instructions and optimization goals can be flexibly adjusted according to different construction needs and conditions to enhance the adaptability and flexibility of the system. By accurately controlling the construction position and reducing deviations, the construction quality can be improved to ensure that the construction results meet the expected requirements.

[0087] In a preferred embodiment of the present invention, detecting whether the construction quality assessment result meets the preset standard, and sending an alarm signal when the construction quality assessment result does not meet the preset standard, may include: Compare the construction quality assessment results with the preset standards to obtain comparative results; Analyze the comparison results to determine whether the construction quality assessment results meet the preset standards. If the assessment results meet the standards, continue monitoring; if not, enter the alarm signal processing process; In the alarm signal processing process, an alarm signal is generated, including log record, email notification, SMS reminder, sound alarm or light indication.

[0088] In an embodiment of the present invention, real-time construction quality assessment data is obtained from a construction quality detection system or related sensors, and these data include indicators such as dimensional accuracy, material strength, and surface flatness. Preset construction quality standards are read from internal storage or external databases, and these standards are formulated according to engineering specifications, design requirements, or industry standards. The obtained construction quality assessment results are compared with the preset standards one by one, and the difference values ​​or similarities are calculated to form a comparison result. The comparison results are analyzed in detail, which may include statistical distribution of difference values, calculation of excess proportions, identification of major sources of deviations, etc. According to the analysis results, it is judged whether the construction quality assessment results meet the preset standards as a whole. This involves setting a threshold or tolerance range, and when the assessment results are within this range, the standards are considered to be met. If the assessment results meet the standards, the construction quality is continuously monitored, the assessment results are updated in real time, and the comparison and judgment process is repeated. If the assessment results do not meet the standards, the alarm signal processing process is entered, and an alarm log is first generated to record detailed information such as the alarm time, the alarm reason, and the difference between the assessment results and the standards. An alarm email is sent to a preset recipient list via an email server, and the email content contains the alarm details and possible recommended measures. Through the SMS gateway, an alarm text message is sent to the mobile phones of relevant personnel, briefly and clearly informing them of the alarm situation and the degree of urgency.

[0089] The built-in audio device can emit a preset alarm sound to attract the attention and immediate response of the on-site personnel. The on-site alarm light or indicator light can be controlled to flash or change color to intuitively display the alarm status.

[0090] Suppose you are monitoring the construction quality of a concrete pour.

[0091] The concrete strength assessment results are obtained from the sensor, and the preset strength standards are read from the database. The assessment results are compared with the standards, and it is found that the concrete strength of some areas is lower than the standard value. The comparison results are analyzed to confirm the proportion and severity of the areas with insufficient strength.

[0092] Since the evaluation results do not meet the standards, it is decided to enter the alarm signal processing process: Generate an alarm log to record the specific circumstances and times of insufficient strength.

[0093] Send email notifications to engineering managers and quality control departments.

[0094] Send SMS reminders to on-site construction personnel, requiring them to check and take measures immediately.

[0095] Trigger an audible alarm to draw the attention of on-site personnel.

[0096] Activate the light indication to mark areas of insufficient intensity with red light.

[0097] By comparing and analyzing the construction quality assessment results with the preset standards in real time, problems in construction can be discovered in time to prevent them from expanding or evolving into more serious quality accidents. Alarm signals are automatically generated and relevant personnel are notified in a variety of ways, greatly improving the response speed and processing efficiency of problems. The generated alarm log records the alarm situation and processing process in detail. Timely alarms and effective processing help ensure that the construction quality meets the preset standards and improve the overall project quality and reliability. Automatic monitoring and alarms reduce the burden of manual monitoring.

[0098] In a preferred embodiment of the present invention, the construction quality assessment results and alarm information are displayed, and remote control is supported, which may include: In an embodiment of the present invention, construction quality assessment data is obtained in real time from a construction quality detection system or related sensors, and these data include but are not limited to specific indicators such as dimensional accuracy, material strength, and surface flatness. The acquired data is preprocessed, such as cleaning, denoising, formatting, etc., to ensure the accuracy and consistency of the data. A user-friendly interface or dashboard is designed to display the construction quality assessment results. The interface includes various forms such as charts, tables, and progress bars to intuitively display the status of each assessment indicator. Determine the layout, color, font and other visual elements of the display content to ensure that the information is clear and easy to read. The processed assessment data is updated to the display interface in real time to ensure that users can view the latest construction quality status at any time. A dynamic data refresh mechanism is implemented, such as timed refresh, event-triggered refresh, etc., to ensure the real-time nature of the display content. When the construction quality assessment result does not meet the preset standard, an alarm message is generated according to the preset alarm rules. The alarm information includes details such as the alarm time, alarm reason, and recommended measures.

[0099] Design a special alarm area or pop-up window on the display interface to highlight the alarm information. The alarm area may use eye-catching colors, icons or animation effects to attract the user's attention. Push the generated alarm information to the display interface in real time to ensure that users can obtain the alarm information as soon as possible. Implement hierarchical management of alarm information, such as displaying and pushing different levels according to the severity or urgency of the alarm. Design a secure remote control interface to allow users to control the construction quality monitoring system through remote devices (such as computers, mobile phones, tablets, etc.). Ensure the security of the interface, such as using encrypted transmission, identity authentication and other mechanisms to prevent unauthorized access and operation, and implement specific functions of remote control, such as adjusting monitoring parameters, setting alarm rules, and viewing historical data.

[0100] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. The prefabricated building construction quality monitoring system based on the Internet of Things is characterized by: include: The sensor module is used to obtain various quality-related data during the construction process in real time, including the installation position of structural parts, connection strength, concrete pouring quality, and ambient temperature and humidity; A data transmission module, used to transmit various quality-related data of the sensor module to the data processing module; A data processing module is used to receive and process data from the data transmission module, perform data analysis, anomaly detection and quality assessment, and generate construction quality assessment results; Intelligent equipment control module, used to automatically adjust the working parameters of construction equipment or stop the operation of faulty equipment according to the construction quality assessment results; An alarm module is used to detect whether the construction quality assessment result meets the preset standard, and send out an alarm signal when the construction quality assessment result does not meet the preset standard; The user terminal module is used to display construction quality assessment results and alarm information, and supports remote control.

2. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 1 is characterized in that: Receive and process data from the data transmission module, perform data analysis, anomaly detection and quality assessment, and generate construction quality assessment results, including: Extract features related to construction quality assessment from the data in the data transmission module, including the installation position deviation of structural parts, the value of connection strength, the uniformity index of concrete pouring, and the range of changes in ambient temperature and humidity; According to the preset anomaly detection threshold, the extracted features are detected, the existing anomalies or abnormal patterns are identified and marked, and anomaly detection results are formed; Based on historical data, a construction quality assessment model is constructed, and the extracted features and feature sets of anomaly detection results are input into the quality assessment model for calculation and reasoning to obtain the construction quality assessment results.

3. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 2 is characterized in that: According to the preset anomaly detection threshold, the extracted features are detected, the existing anomalies or abnormal patterns are identified and marked, and the anomaly detection results are formed, including: The preset anomaly detection model is trained and learned through the isolation forest algorithm to obtain a trained anomaly detection model; The extracted features, including the installation position deviation of the structural parts, the numerical value of the connection strength, the uniformity index of the concrete pouring, and the ambient temperature and humidity, are passed as input data to the trained anomaly detection model; After training, the anomaly detection model analyzes the input features and calculates an anomaly score for each data point; According to the anomaly score, a threshold is set, and the anomaly score of each data point is compared with the threshold. When the anomaly score of a data point exceeds the threshold, the data point is marked as an anomaly point or an abnormal pattern; The data points marked as abnormal are aggregated to form the anomaly detection results, including the feature values, occurrence time and abnormal type information of the abnormal points.

4. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 3 is characterized in that: The anomaly score of a data point is calculated as: ; in, Represents data points Anomaly score of Indicates the size of the data set; , represents the scaling factor; represents the number of trees in the isolation forest; Represents the index of the tree; Represents data points In the The path length in the tree; Indicates The weight of each tree; Indicates Tree-specific adjustment factor; represents a positive number; Indicates adjustment The coefficient of influence; Indicates the offset; Represents parameters related to tree depth.

5. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 4 is characterized in that: Based on historical data, a construction quality assessment model is constructed, and the extracted features and feature sets of anomaly detection results are input into the quality assessment model for calculation and reasoning to obtain the construction quality assessment results, including: Extract construction quality-related features from historical construction data, including installation position deviation of structural components, connection strength, concrete pouring uniformity, and ambient temperature and humidity; The random forest model is used as the construction quality assessment model, and the construction quality assessment model is trained using the training set in the historical data. During the training process, the parameters and hyperparameters of the construction quality assessment model are adjusted to obtain a trained quality assessment model. The extracted features and feature sets of anomaly detection results are input into the trained construction quality assessment model for analysis, and the construction quality assessment results are obtained through calculation, including the construction quality score, qualified or unqualified judgment, and identification of potential problems.

6. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 5 is characterized in that: The calculation formula for the construction quality score is: ； in, A score indicating the quality of construction; represents a constant; Indicates the total number of features; , The index representing the feature; Indicates The value of a feature; Indicates The type coefficient of each feature; Indicates The standardized coefficients of the features; Indicates The standardized coefficients of the features; represents the adjustment parameter; represents the base of natural logarithms; Indicates The minimum value of the eigenvalue among all samples; Indicates The maximum value of the eigenvalue among all samples; Indicates and The interaction coefficient between the features; Indicates The value of a feature; Indicates time; Indicates time The weight of Indicates time The characteristic value when ; Represents the time decay rate.

7. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 6 is characterized in that: According to the construction quality assessment results, the working parameters of the construction equipment are automatically adjusted or the faulty equipment is stopped, including: Presetting a threshold for construction quality assessment, and comparing the actual construction quality assessment result with the threshold to obtain a comparison result between the assessment result and the threshold; According to the comparison between the evaluation result and the threshold value, it is determined whether to adjust the equipment working parameters or stop the faulty equipment; For equipment that needs to adjust working parameters, the adjustment plan is determined based on the comparison between the evaluation results and the threshold, including the adjustment of speed, force and temperature parameters; for faulty equipment, the equipment stops working immediately; Generate corresponding control instructions based on the decision to adjust the plan or stop work, and send the control instructions to the construction equipment; The construction equipment automatically adjusts its own working parameters to the specified values ​​or stops working immediately according to the control instructions, and starts the fault handling process, including alarm and self-diagnosis.

8. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 7 is characterized in that: According to the decision to adjust the plan or stop work, generate corresponding control instructions and send them to the construction equipment, including: receiving a decision signal to adjust the plan or stop working, and determining an initial radius of the trust region; Define an objective function to evaluate the quality of control instructions; Searching for new control instructions in the trust region and evaluating the impact of the new control instructions on the objective function to obtain an evaluation result; Compare the new control instruction with the current control instruction to obtain a comparison result, and determine whether to update the control instruction according to the comparison result; Continuously search for new control instructions in the trust domain, evaluate each new instruction, and determine whether to update the current control instruction based on the evaluation results until the maximum number of iterations is reached, generate the final control instruction, and send the final control instruction to the construction equipment.

9. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 8 is characterized in that: The calculation formula of the objective function is: ； in, represents the objective function; , represents the weight coefficient; Indicates the actual construction location; Indicates the expected construction location; Indicates the base tolerance; Indicates the tolerance adjustment factor; Indicates basic working power; Indicates device parameters Linear influence coefficient on power; Indicates equipment parameters that affect energy consumption; Indicates device parameters The secondary influence coefficient on power; Indicates base time; Indicates the amplitude of periodic changes in energy consumption; represents the angular frequency; Indicates time.

10. The Internet of Things-based prefabricated building construction quality monitoring system according to claim 9, characterized in that: Check whether the construction quality assessment results meet the preset standards. When the construction quality assessment results do not meet the preset standards, an alarm signal is issued, including: Compare the construction quality assessment results with the preset standards to obtain comparative results; Analyze the comparison results to determine whether the construction quality assessment results meet the preset standards. If the assessment results meet the standards, continue monitoring; if not, enter the alarm signal processing process; In the alarm signal processing process, an alarm signal is generated, including log record, email notification, SMS reminder, sound alarm or light indication.

Images